The abbreviated injury scale as a predictor of outcome of severe head injury
A. D. Walder
Queens Medical CentreUniversity Department of Anaesthesia
P. M. Yeoman
Department of Intensive Care, Adult Intensive Care UnitQueens Medical Centre
Department of RadiologyDerby Royal Infirmary
Cite this article as:
Walder, A.D., Yeoman, P.M. & Turnbull, A. Intensive Care Med (1995) 21: 606. doi:10.1007/BF01700170
This study examined the correlation between the worst abbreviated injury scale 1990 (AIS) intracranial severity score and outcome following severe head injury.
The initial CT scans of 109 severly head injury patients were examined by a neuroradiologist and classified according to the worst applicable intracranial severity code from the AIS. This score was then correlated with the glasgow outcome scale (GOS) at 6 months. For comparison, the GOS was also correlated with the diffuse injury scale (DIS) described by L.F. Marshall et al. , the worst post-resuscitation Glasgow coma score (GCS) in the first 24 h, and the head injury outcome prediction tree described by Choi et al. .
Our results show Spearman rank correlation coefficients of 0.58 (p<0.001), 0.47 (p<0.001), 0.45 (p<0.001), and 0.31 (p<0.01) for the correlation between the AIS, prediction tree, DIS, and GCS respectively and the GOS. Independent outcome (i.e. GOS good or moderate) was strongly predicted by an AIS of 3 or less (positive predictive value 95%, specificity 98%, sensitivity 40%, likelihood ratio 25∶1). Death or vegetative survival was less strongly predicted in patients with an AIS of 5 (positive predictive value 71%, specificity 75%, sensitivity 67%, likelihood ratio 2.7∶1).
The AIS, based on initial CT scan, provides useful prognostic information in patients with severe head injury.
Severe head injuryOutcomeComputerised tomographic scanAbbreviated injury scale